While dementia carries profound implications for human suffering and economic burdens among more than 5 million people in the US and their caregivers, these implications are magnified in low-income communities. Dementia care management - in which social workers or nurses conduct structured assessments and work with patients and caregivers to identify problems and execute individually-tailored management protocols on an ongoing basis - has been proven to improve quality and outcomes of care when delivered in a clinic or as a combination of home visits and telephone. However, with growing cost-containment pressures, healthcare systems are increasingly pressured to deliver care management less expensively, i.e, exclusively by telephone. Yet, in Latino immigrant communities, cultural, language and educational barriers may mean that more intensive and in-person interactions at home and in community settings are needed to engage care recipients and caregivers in order to benefit from care management. In this randomized trial, two different strategies for delivery of dementia care management protocols will be compared. A community-centered care management strategy will partner a healthcare organization care manager with one from the Alzheimer's Association, who will collaborate in providing care management through home visits and in-person meetings at local facilities. In the comparison strategy, a single care manager at the healthcare organization will provide protocol-based telephone care management for the same number of participants. 250 patient/caregiver dyads from an economically-impoverished, predominantly Latino community in Los Angeles will be randomized and followed for 12 months. Aims are: (1) to test whether a community-centered dementia care management strategy compared to a telephone-only care management strategy yields lower caregiver burden and fewer challenging behaviors in care recipients. We hypothesize that the community-centered strategy will produce better outcomes due to greater cultural grounding from in-person delivery of care management in community settings or at home;(2) to test whether retention of participant dyads will be greater with a community- centered or a telephone-only care management strategy. We hypothesize that community-centered care management will be more successful in retention;and (3) to compare the costs and conduct a cost- effectiveness analysis of the two strategies. We hypothesize that the community-centered strategy will cost more but will produce greater reductions in caregiver burden and in problem behaviors in the care recipient. Findings will provide new knowledge on relative effectiveness of dementia care management strategies and their costs, in a predominantly Latino community in the public safety net system. It will measure caregiver burden and problem behaviors in care recipients, both of which are linked to costly institutionalization. With dementia's growing burden, these data will be critically important to administrators in public safety net settings, for decisions about the relative value of care delivery strategies in a highly-constrained fiscal climate. PUBLIC HEALTH RELEVANCE: Dementia is a condition that is growing in prevalence and which harms not only the afflicted individual but also adversely affects the health of their family and other informal caregivers. New methods for delivering comprehensive assistance to persons with dementia and their caregivers are known to be effective and can delay nursing home placement, but this study will discover (1) whether more face-to-face involvement rather than telephone delivery of this assistance will work better among poor patients in Los Angeles, and (2) if one method is better than the other, what are the differences in costs between them. These data will enable administrators in public health care settings around the US and non-profit foundations addressing dementia patient and caregiver need to decide what method provides the best value and the best outcome relative to its cost.